In traditional surround sound systems, a listener places 5 or more speakers at various positions around a listening position (sometimes also referred to as a listening area) to create an immersive sound experience for a listener. Each of the speakers in the system typically receives its own audio signal from an audio source, and consequently, the listener typically must wire each of the speakers to the audio source. The speakers in the audio system then produce sound that converges at the listening position to properly create a surround sound experience for the listener.
Virtual surround is a surround sound technique that can make sound appear to come from locations other than the location of the actual speakers in order to create a surround sound experience for a listener. As a result, virtual surround sound systems typically use fewer speakers than traditional surround sound systems, and the speakers in a virtual surround sound system are usually located in front of the listener. Virtual surround sound systems are thus more practical for a variety of different setups, such as with a personal computing system or a television.
Virtual surround sound widens the soundscape beyond the physical location of the speakers used to produce the sound, and is based on how humans localize sound. Humans localize sound using three methods: 1) Interaural Intensity Difference (IID), 2) Interaural Time Difference (ITD), and 3) Spectrally, with the Head Related Transfer Function (HRTF). Interaural Intensity Difference occurs when a sound is louder at one ear than at the other ear. This can occur when the sound source is closer to one of the ears. Similarly, Interaural Time Difference occurs when the sound reaches one ear before it reaches the other ear because the sound source is closer to one of the ears. This can cause a difference in time and therefore a difference in phase between the ears. A Head Related Transfer Function refers to the unique spectral shaping of sound as it reflects off of the pinna (outer ear), head, and shoulders of the listener. The spectral shaping can vary depending on the location of the sound source. Additionally, the spectral shaping can vary depending on the particular listener.
Virtual surround sound may employ one or more different techniques to create the impression on a listener that sound is coming from a location other than the location of the speakers based on one or more of the three above methods. For example, dipole beamforming is one method for creating virtual surround using IID. Dipole pairs of transducers can be used to artificially increase the difference in sound level between the ears. The transducers in a dipole pair are driven out of phase with each other in order to create a null for certain frequencies or channels, and a delay is used to steer the radial direction of the null. The result is that sound for certain frequencies or channels is more intense at one ear of the listener compared to the other ear, and the listener is left with the impression that the sound is originating from a location other than the actual location of the transducers producing the sound.
The transducers used in a dipole beamforming array are generally chosen for their dispersion characteristics in the targeted array frequency range. For example, woofers have good efficiency and near omni-directional radiation at lower frequencies. Woofers thus are a good choice for a lower frequency array. At higher frequencies, woofers start to beam and have less consistent directionality. This phenomenon is related to the size of the transducer relative to the wavelength of sound that it produces. In contrast, tweeters are physically smaller and thus have better dispersion for higher frequencies with smaller wavelengths. Therefore, tweeters are a good choice for a high frequency array. However, higher frequencies can be difficult to properly implement with a dipole beamforming array. Ideally, the distance between the center of the transducers used to form a dipole pair in a dipole beamforming system should be separated by a quarter-wavelength; the relative wavelength in this case is the center frequency for which the dipole pair is optimized. Since higher frequencies have smaller wavelengths, it may not always be physically possible to place tweeters (or other transducers) close enough together for an optimized dipole beamforming system. The array is optimized over approximately 4 octaves: 2 octaves above and below the center frequency. Above this frequency range, the distance between the transducers can become large relative to the wavelength of sound being produced, and more nulls are created as the frequency increases. The implication of this is that the sound at one ear may no longer be louder than at the other ear, and the virtual surround effect is lost. Below this range, the efficiency of sound production can decrease as sound from the out of phase transducers cancels.
Accordingly, it would be desirable to have a better virtual surround system that produces constant directivity across a wide range of frequencies in a small system that is useful for a variety of different setups. A number of different techniques are known in the art for creating virtual surround sound. For example, U.S. Application Pub. No. 2006/0072773 entitled “Dipole and monopole surround sound speaker system,” U.S. Application Pub. No. 2009/0060237 entitled “Array Speaker System,” U.S. Application Pub. No. 2008/0273721 entitled “Method for spatially processing multichannel signals, processing module, and virtual surround-sound systems,” and U.S. Application Pub. No. 2003/0021423 entitled “System for transitioning from stereo to simulated surround sound” all show different virtual surround systems. However, each of these systems could be improved to have more constant directivity across a wider range of frequencies.